Radio devices having an analog front-end (AFE) undergo extensive calibrations and tests in the manufacturing environment after production by utilizing a radio-frequency (RF) tester to check whether performance the device is within specification and/or to retune certain components. However, since most calibration algorithms heavily rely on off-chip RF equipment or number crunchers, in-the-field monitoring, calibration, reconfiguration, device optimization and/or re-tuning of the analog front-end is difficult or not able to be performed. Furthermore, post-manufacturing test times become longer, for example greater than one minute, and the testing cost per unit may become a significant portion of the total manufacturing cost of the radio analog front end. Testing time inevitably will increase in the future as radio devices evolve towards smaller technologies having more variations and/or more complex radios, for example radio devices implementing multiple-input, multiple-output (MIMO), multiband radios, and so on. A few calibrations and/or tests are currently executed on the chip to save cost, however current implementations are done only for individuals tests and in an ad-hoc manner. Such an approach not only represents a huge waste of chip area via circuit duplication, but also complicates the manageability and development time, thereby increasing time to market (TTM). Post tape out bug fixing also becomes difficult if not nearly impossible. Furthermore, computationally complex calibration strategies currently cannot be executed on chip because current chips lack the flexibility, speed and/or computational resources to accomplish this. For example, a spectrum (blocker) sensing algorithm may involve one billion Fast Fourier Transform (FFT) butterfly operations per sec, which is impossible to realize with current multiply and accumulate (MAC) processors unless a huge amount of memory and/or dedicated digital signal processing logic is added requiring increased area, cost, and/or manageability.
After some time, a fraction of deployed chips may fail or start to fail. Since self-testing is almost inexistent for radios, the chip cannot be diagnosed remotely, and the device has to be shipped back to the original equipment manufacturer (OEM). Since the OEM typically may not have the expensive test equipment and expertise to do analog and radio-frequency (RF) tests, the only available option for the OEM is to replace the chip. However, replacement results in an unnecessarily large cost for the OEM due to shipping the device or replacing chips that might not be broken. Additionally, when a chip is untested and merely replaced, the chip manufacturer does not get any diagnosis of exactly what part of the chip caused the failure and whether the problem could have been solved by recalibration and/or retuning.
Currently every commercial radio platform which will be sold has to be certified by regulatory agency such as the Federal Communications Commission (FCC), industry interoperability groups, standards groups, and so on. The different organizations check whether the radio performs conformance to their specifications, for example spectrum mask compliance, sensitivity, transmit power, and so on. This is accomplished by performing extensive external measurements on a few samples of the new platform carrying the radio chips. Every time something is changed to the design or firmware of the radio chip or platform, the device has to be recertified. The certification and recertification process takes up to six months or so, thereby significantly delaying the time to market (TTM). Since there is an emerging trend to integrate the analog radio and the digital baseband processor on the same chip, and further to integrate the radio on the main processor die, there are potentially severe implications for the certification of the radios because recertification will be required every time something is altered in the overall chip design even when the change has little to do with the radio itself. Thus, the certification process may present a significant delay in designing a processor chip that incorporates a radio on the chip. Another trend is the shift towards smaller and smaller silicon technologies for implementing the radio chips which may result in more on-die variations from chip to chip. Testing and certifying only a limited number of chip samples during the certification phase ultimately may become insufficient to statistically account for the chip to chip variation.
It will be appreciated that for simplicity and/or clarity of illustration, elements illustrated in the figures have not necessarily been drawn to scale. For example, the dimensions of some of the elements may be exaggerated relative to other elements for clarity. Further, if considered appropriate, reference numerals have been repeated among the figures to indicate corresponding and/or analogous elements.